The Science (and Practice) of Using Behavioural …...• Feng, Bing, Jima Oyunsuren, Mykyta Tymko,...

The Science (and

Practice) of Using

Behavioural Insights

@dilipsoman

Structure of remarks

• Claims and observations

• Psychological considerations

• Plumbing problem

• Using Behavioural Insights

Everyone is in the Business of Changing Behaviour…

COMPLIANCESWITCHING /

PURCHASING

CONSUMPTION ACCELERATION

Econs!

•

Humans!

Design for Humans, not for Econs

•

The Four – “ition”s

Motivation Cognition

Perception Emotion

Context Procrastination Inertia

Three segments in Behaviour Change!

Motivated

Enthusiasts

Diehard

Opponents

Naïve

Intenders

“Done” “No way” Yes, I’ll do it tomorrow”

Source: Soman and Ly (2018), The Growing Market for Self-Control

The Behaviour Change Puzzle

Source: Soman (2015), The Last Mile

Behaviour Change as a Plumbing Problem

Source: Soman (2015), The Last Mile

Behavioural Scientist as Strategist

(overall organizational operations)

Behavioural Scientist as Designer

(policy and program design)

Behavioural Scientist as Problem Solver

(last mile challenges)

Behavioural Scientist as Auditor

(proactive problem identification)

The Roles of BI

Source: Feng at al. (2018)

The Three Legged Stool for Success

Behavioural science (and understanding of human psychology)

& empathy. Behavioural engineering (choice architecture and

the approaches Data & empirical science (analytics,

experimental design, and ability to analyze and interpret

experiments).

An organizational structure and processes that allow for quick

feedback loops to be incorporated, and for the organization to be

able to change course so that a test-learn-adapt strategy can be

put in place.

A mindset that data (rather than theory or a pre-committed

course of action) drive decision making in the organization. This

mindset calls for an empathic mindset at the intersection of BI and

design thinking, and a relentless desire to test using experiments

through RCT’s or other methodologies.

EXPERTISE

AGILITY

EMPIRICISM

The Costs of Experimentation

Investing in hard and soft infrastructure • Online experimental platforms that reduce time to collect data, Participant pools that recruit from the appropriate

populations, Laboratory space and facilities to collect behavioural data, Researchers with expertise in experimental

methods

Appropriate problem selection • Decomposing large challenges into precise behaviours, Building horizontally scalable behaviour change problems,

Studying outcomes with short outcome cycles / feedback loops

Building whatworks databases • Allows organizations to narrow down possible interventions to test

Ethics in experimentation • Ensuring that interventions a) are consistent with users values and interest, b) are for legitimate ends, c) do not violate

rights, d) are transparent and e) do not take things away without consent

Reframing failure and Incentivizing Experimentation • Realizing and clearly communicating that a null-result is a learning opportunity will minimize the fear of failure. Explicitly

rewarding experimentation will further build the test-learn-adapt culture

How Do I Reduce the Costs of Experimentation?

• Process frameworks versus prescriptive frameworks

• Descriptive versus prescriptive frameworks

• Work well: Baseline, training, marketing

• Pitfalls: Misapplication, glossing over the details of the “pipeline,” misallocation of resources, using framework as a checklist

Beyond frameworks

Source: BIT

Source: BIAS project, MDRC

Source: Soman (2017), from BIAS Annual Report

In Conclusion

• Holistic approach to behaviour change

• Behaviour change as a plumbing

problem

• Going beyond the last mile

• Reducing the costs of experimentation

• Beware of the Elegant Simplicity (and

Pitfalls) of Simple Frameworks

Resources

www.rotman.utoronto.ca/bear

@UofT_BEAR @dilipsoman @BE101x

• Soman, D. (2015). The Last Mile: Creating Social and Economic Value from Behavioural Insights. Toronto:

University of Toronto Press.

• Feng, Bing, Jima Oyunsuren, Mykyta Tymko, Melanie Kim and Dilip Soman (2018), How Should Organizations

Best Embed and Harness Behavioural Insights? A Playbook, Toronto, Canada: Behavioural Economics in Action

at Rotman (BEAR) Report series, available at http://www.rotman.utoronto.ca/bear

• Behavioural Economics in Action (BE101x) MOOC: https://www.edx.org/course/behavioural-economics-in-action

• All photographic images have been used under the creative commons licensing arrangement. Illustrations are ©

BEAR. Details on sources are available on request.